2,4,6-triarylphosphinines versus 2,4,6-triarylpyridines: An investigation of the differences in reactivity between structurally related aromatic phosphorus and nitrogen heterocycles

Article abstract of DOI:10.1002/chem.201302441

The novel atropisomeric pyridine derivative rac-10 has been synthesized and structurally characterized. In contrast to its phosphorus analogue 3, axially chiral 10 has a considerably lower rotational barrier as estimated by DFT calculations. However, the presence of the two enantiomers could be confirmed by means of chiral analytical HPLC analysis and by protonation experiments with a chiral acid. Compound rac-10 could be further dehydrogenated by treatment with DDQ to the benzo(h)quinoline derivative rac-12. This conversion failed for the phosphorus analogue rac-3. Interestingly, although 2,4,6-triarylphosphinines undergo facile C-H activation with [CpIrCl₂]₂ in the presence of NaOAc, this reaction does not proceed with the corresponding pyridine derivatives. On the other hand, the latter ones can be selectively ortho-metalated with Pd(OAc)₂, leading to acetate-bridged dimeric species, which could be unambiguously confirmed by means of X-ray crystal structure analysis. The treatment of phosphinines with Pd(OAc)₂ led instead to the formation of the unusual cofacial oxidative coupling products 16 and 17, which consist of a phosphorus-containing cage structure.

Full text of DOI:10.1002/chem.201302441

DOI: 10.1002/chem.201302441
2,4,6-Triarylphosphinines versus 2,4,6-Triarylpyridines: An Investigation of
the Differences in Reactivity between Structurally Related Aromatic
Phosphorus and Nitrogen Heterocycles
Jarno J. M. Weemers,[a] Jelena Wiecko,[b] Evgeny A. Pidko,[a] Manuela Weber,[b]
Martin Lutz,[c] and Christian Mꢀller*[a, b]
Abstract: The novel atropisomeric pyr-
idine derivative rac-10 has been synthe-
sized and structurally characterized. In
contrast to its phosphorus analogue 3,
axially chiral 10 has a considerably
lower rotational barrier as estimated
by DFT calculations. However, the
presence of the two enantiomers could
be confirmed by means of chiral ana-
lytical HPLC analysis and by protona-
tion experiments with a chiral acid.
Compound rac-10 could be further de-
hydrogenated by treatment with DDQ
to the benzo(h)quinoline derivative
rac-12. This conversion failed for the
phosphorus analogue rac-3. Interesting-
ly, although 2,4,6-triarylphosphinines
the other hand, the latter ones can be
selectively ortho-metalated with Pd-
AHCTUNGRTEN(GNUN OAc)2, leading to acetate-bridged di-
meric species, which could be unambig-
uously confirmed by means of X-ray
crystal structure analysis. The treat-
À
undergo facile C H activation with
[Cp*IrCl2]2 in the presence of NaOAc,
this reaction does not proceed with the
corresponding pyridine derivatives. On
ment of phosphinines with PdACHTUNGTRENNUNG(OAc)2
led instead to the formation of the un-
usual cofacial oxidative coupling prod-
ucts 16 and 17, which consist of a phos-
phorus-containing cage structure.
Keywords: coordination chemistry ·
heterocycles · phosphinines · pyri-
dines
Introduction
otherwise identical compounds, which differ only in the
nature of the heteroatom (Figure 1).[3,4]
l3-Phosphinines are low-coordinate phosphorus compounds
that possess not only significantly different steric and elec-
tronic properties compared to classical trivalent phosphorus
compounds, but also compared to their nitrogen analogues,
the pyridines.[1,2]
We have recently started to investigate extensively the
preparation of functionalized phosphinines via the pyrylium
salt route, originally described by Mꢀrkl for the synthesis of
2,4,6-triphenylphosphinine (1).[1] This modular procedure al-
lowed us to introduce substituents into specific positions of
the heterocyclic framework; a feature which is essential for
ligand design and their potential use in more applied re-
search fields. Moreover, the pyrylium salt route offers at the
same time the synthesis of the corresponding pyridines, such
as 2, having the same substitution pattern as their phospho-
rus analogues. In this way, it is possible to directly compare
Figure 1. 2,4,6-Triphenylphosphinine (1), 2,4,6-triphenylpyridine (2) and
atropisomeric phosphinine 3 (only one enantiomer is shown).
We have extended this synthetic procedure also to the
preparation of the first configurationally stable atropisomer-
ic phosphinine 3-Sa and 3-Ra, by locating a sterically de-
À
manding group in 6-position and an additional CH3 group
in 5-position of a rigid phosphorus-containing heterocyclic
framework, starting from the corresponding functionalized
propiophenone derivative (Figure 1).[5] Consequently, we
achieved axial chirality with a sufficiently high energy barri-
er for internal rotation to prevent enantiomerization. Both
enantiomers could be isolated by means of chiral analytical
HPLC and their absolute configurations could be assigned
by comparing the experimentally determined CD spectra
[a] J. J. M. Weemers, Dr. E. A. Pidko, Prof. Dr. C. Mꢁller
Chemical Engineering and Chemistry
Eindhoven University of Technology
Den Dolech 2, 5600 MB Eindhoven (The Netherlands)
[b] Dr. J. Wiecko, M. Weber, Prof. Dr. C. Mꢁller
Institut fꢁr Chemie und Biochemie - Anorganische Chemie
Freie Universitꢀt Berlin, Fabeckstr. 34/36, 14195 Berlin (Germany)
E-mail: c.mueller@fu-berlin.de
À
with the theoretically calculated ones. C H activation at
[c] Dr. M. Lutz
room temperature on the adjacent aryl-substituent in 2-posi-
Bijvoet Center for Biomolecular Research, Utrecht University
Padualaan 8, 3584 CH Utrecht (The Netherlands)
tion of the phosphorus heterocycles was achieved for 1 and
14458
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 14458 – 14469

FULL PAPER
Figure 2. IrIII complexes 4 and 5 containing cyclometalated phosphinines
1 and 3.
rac-3 using [Cp*IrCl2]2 as metal precursor in the presence of
NaOAc and the products (4, 5) could be characterized by
single crystal X-ray diffraction (Figure 2).[5–7]
Scheme 1. Synthesis of 1,5-diketone 8 and pyrylium salt rac-9. Reaction
conditions: a) tetrabutylammonium iodide, KOH (10% aq. solution),
DMSO;
b) trans-benzylideneacetophenone
(1 equiv),
BF3·Et2O,
Inspired by these results we started to investigate the de-
velopment of analogous systems based on nitrogen, as chiral
pyridine derivatives are generally of high interest as chiral
building blocks not only for natural product synthesis but
also for applications in asymmetric catalysis. In order to
obtain a closer insight in the differences of aromatic phos-
phorus and nitrogen heterocycles, which have otherwise
identical structures, we set out to further compare the
reactivity of such compounds towards IrIII and PdII metal
precursors.
ClCH2CH2Cl, 708C, overnight.
Scheme 2. Synthesis of atropisomeric pyridine rac-10 starting from pyryli-
um salt rac-9. Reaction conditions: a) ammonia (25% aq. solution),
EtOH, reflux.
Results and Discussion
Propiophenone 6 is the key compound for introducing axial
chirality in the final phosphinine 3 (Figure 1). Moreover, we
have recently shown that it is necessary to also introduce
substituents in 3-position of the heterocycle and to create
more rigidity to the framework using the tetralone-based
chalcone 7 in order to prevent the formation of side prod-
ucts during the synthesis of the pyrylium salt derivative.
Since it is impossible to prepare pyrylium salt rac-9 directly
from 6 and 7, the 1,5-diketone 8 was first synthesized by fast
treatment of 6 with 7 in a mixture of DMSO and 10% aque-
ous KOH solution and in the presence of tetrabutylammoni-
um iodide. Subsequent treatment with 1 equiv of benzyli-
dene-2-acetophenone in the presence of BF3·Et2O at 708C
afforded the racemic pyrylium salt rac-9 as a yellow solid in
almost quantitative yield (Scheme 1).[5]
The synthesis of phosphinines from pyrylium salts is gen-
erally far from straightforward and every reaction needs its
specific reaction conditions. On the other hand, the synthesis
of pyridines from pyrylium salts is often much more facile.
However, it turned out that rac-10 could not be obtained by
bubbling anhydrous ammonia through a solution of rac-9 in
2-methyltetrahydrofuran at 508C, which is generally the re-
action of choice for the conversion of triarylpyrylium salts
into their pyridine counterparts.[4] Nevertheless, rac-9 could
easily be converted to the corresponding atropisomeric pyri-
dine rac-10 by treatment with 25% aqueous ammonia in
EtOH at 208C in 81% yield (Scheme 2).[8]
In order to obtain insight into the configurational stability
and to estimate the energy barrier for internal rotation of
the atropisomeric pyridine derivative 10 we performed DFT
calculations.[9,10] Much to our surprise it turned out that the
predicted rotational barrier of DE° =111 kJmolÀ1 is about
50 kJmolÀ1 less than the energy barrier calculated for phos-
phinine 3 (Figure 3, top).
This result suggests that enantiomerically pure 10 tends to
racemize slowly at lower temperature, whereas enantiomeri-
cally pure phosphinine 3 is configurationally stable at least
up to 1408C. We could show that a racemic mixture of 3 can
consequently be separated experimentally into the corre-
sponding enantiomers.
The reason for this rather large difference in the rotation-
al barriers determined for the two structurally related het-
erocycles might be due to the pronounced differences that
exist between the two heteroatoms. In fact, the phosphinine
heterocycle is best described as a distorted hexagon due to
À
À
the larger P Ca bond distances compared to the N Ca bond
lengths in pyridine (Figure 4). As a result, the aryl groups in
ortho-position of the phosphinine framework are slightly
bent away from the phosphorus atom, which obviously has
an influence on the steric interaction between the CH3
group in 5-position of the heterocycle and the substituted
naphthyl group. On top of that, the phosphorus lone pair of
the phosphinine is more diffuse and less directional than the
nitrogen lone pair in pyridines.[11] Also in this case a steric
Chem. Eur. J. 2013, 19, 14458 – 14469
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
14459

C. Mꢁller et al.
ence of a racemate, we attempted to protonate the racemic
mixture of 10 with one enantiomer of a chiral acid. Thus,
treatment of rac-10 with an equimolar amount of S-cam-
phor-10-sulfonic acid in CH2Cl2 at room temperature result-
ed indeed in the formation of 11 in quantitative yield. The
1H NMR spectrum clearly shows the presence of four
methyl resonances in the aliphatic region, which confirmed
that compound 11 was formed as a diastereomeric mixture.
As expected, phosphinine rac-3 cannot be protonated with
S-camphor-10-sulfonic acid due to the low basicity of the
low-coordinate phosphorus atom (Scheme 3).
Scheme 3. Protonation of atropisomeric pyridine rac-10 with S-camphor-
10-sulfonic acid. Reaction conditions: a) CH2Cl2, 208C, t=30 min.
Colorless crystals suitable for X-ray diffraction were ob-
tained by slow evaporation from a concentrated solution of
rac-10 in ethanol. Pyridine rac-10 crystallizes in the mono-
clinic space group P21/n (no.: 14) and the molecular struc-
ture along with selected bond lengths and angles is shown in
Figure 5. From the molecular structure of rac-10 it is obvi-
ous that the naphthyl group in 6-position is located almost
perpendicular to the nitrogen heterocycle with an N(1)-
C(1)-C(15)-C(24) torsion angle of À102.6(2)8, whereas the
bridged phenyl group in 2-position is bent slightly out of the
plane of the nitrogen atom with an N(1)-C(13)-C(12)-C(11)
torsion angle of À22.6(3)8. The bond lengths between the ni-
trogen atom and adjacent carbon atoms in compound rac-10
*
Figure 3. A scan of the potential energy for pyridine 10 ( ) and phosphi-
&
nine 3 ( ) as a function of the torsion angle (top) and chiral HPLC analy-
sis of rac-10 (bottom).
À
À
were found to be N(1) C(1)=1.348(3) ꢃ, N(1) C(13)=
1.345(3) ꢃ, whereas the angle between C(1)-N(1)-C(13) is
118.51(17)8. These values are in the same range as observed
for 2,4,6-triphenyl-substituted pyridines (1.34–1.35 ꢃ and
118–1198, respectively).[12] In contrast, the remaining
À
carbon carbon bonds in the nitrogen heterocycle are some-
what longer compared to their triphenyl-substituted counter-
À
parts (1.375–1.386 ꢃ) and were found to be C(1) C(2)=
À
À
1.399(3) ꢃ, C(2) C(3)=1.412(3) ꢃ, C(3) C(4)=1.400(3)
and C(4)-C(13)=1.404(3) ꢃ.
Figure 4. Structural and electronic comparison between heterocycles rac-
3 and rac-10.
We then turned our attention to the dehydrogenation of
rac-10 to its benzo(h)quinoline derivative rac-12. Compound
rac-12 could easily be synthesized in one step by treating
repulsion of the phosphorus lone pair with the substituted
naphthyl group is effective.
Nonetheless, both enantiomers of pyridine 10 could be de-
tected by means of analytical HPLC on a chiral stationary
phase with n-hexane/2-propanol (99:1) as eluent (Chiralpak
rac-10
with
2,3-dichloro-5,6-dicyano-1,4-benzoquinone
(DDQ)[13] in toluene under reflux conditions, and was isolat-
ed as a pink solid in good yield (71%) after column chroma-
tography. Unfortunately, this reaction does not work for the
corresponding phosphinine derivatives under comparable re-
action conditions (Scheme 3), as no product formation could
IC,
t
R1 =7.05 min,
t
R2 =9.15 min,
258C,
flow-rate
1.0 mLminÀ1; Figure 3, bottom). All attempts to separate
the racemate and to determine the rotational barrier experi-
mentally, however, failed. To further demonstrate the pres-
be observed by either H or 31P{1H} NMR spectroscopy.
1
14460
www.chemeurj.org
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 14458 – 14469

2,4,6-Triarylphosphinines versus 2,4,6-Triarylpyridines
FULL PAPER
Figure 6. Molecular structure of rac-12 determined by X-ray crystallogra-
phy (top and side view. Displacement ellipsoids are shown at the 50%
probability level. Only one independent molecule is shown. Selected
Figure 5. Molecular structure of rac-10 determined by X-ray crystallogra-
phy (top and side view). Displacement ellipsoids are shown at the 50%
probability level. Only one enantiomer is shown. Selected bond lengths
À
À
À
bond lengths [ꢃ]: N(1) C(1): 1.340(5), N(1) C(13): 1.357(5), C(1) C(2):
À
À
À
[ꢃ]: N(1) C(1): 1.348(3), N(1) C(13): 1.345(3), C(1) C(2): 1.399(3),
À
À
À
1.407(6), C(2) C(3): 1.383(6), C(3) C(4): 1.435(5), C(4) C(13): 1.401(5);
bond angles [8]: C(1)-N(1)-C(13): 117.8(3); torsion angles [8]: N(1)-C(1)-
C(15)-C(24): 81.2(5), N(1)-C(13)-C(12)-C(11): 2.5(6).
À
À
À
C(2) C(3): 1.412(3), C(3) C(4): 1.400(3), C(4) C(13): 1.404(3); bond
angles [8]: C(1)-N(1)-C(13): 118.51(17); torsion angles [8]: N(1)-C(1)-
C(15)-C(24): À102.6(2), N(1)-C(13)-C(12)-C(11): À22.6(3).
Colorless single crystals suitable for X-ray diffraction
were obtained by slow evaporation from a concentrated so-
lution of rac-12 in ethanol. Pyridine rac-12 crystallizes in the
monoclinic space group P21 with four independent mole-
cules in the asymmetric unit and the molecular structure of
one conformer along with selected bond lengths and angles
is shown in Figure 6.[14] As an example, the molecular struc-
ture of one independent molecule of rac-12 in the asymmet-
ric unit is compared with that of rac-10 (see Figure 5). Simi-
lar to the molecular structure of rac-10, the naphthyl group
in 6-position of rac-12 is almost located perpendicular to the
nitrogen heterocycle with an N(1)-C(1)-C(15)-C(24) torsion
angle of 81.28. This value is comparable to compound rac-
10, whereas the completely aromatic bridged phenyl group
in 2-position is almost in the same plane as the nitrogen
atom with an N(1)-C(13)-C(12)-C(11) torsion angle of only
2.58 (Scheme 4).
Scheme 4. Synthesis of dehydrogenated pyridine rac-12 starting from
atropisomeric pyridine rac-10. Reaction conditions: a) DDQ, toluene,
reflux. The reaction does not work in the case of rac-3.
Chem. Eur. J. 2013, 19, 14458 – 14469
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
14461

C. Mꢁller et al.
The molecular structure of rac-12 in the crystal further re-
and frequently used for cyclometalation reactions as
well.[17,18] The cyclopalladation reactions were carried out
using PdACTHGNUTRENNU(G OAc)2 as the metal precursor. In contrast to the at-
À
vealed that the N(1) C(13) bond length (1.357(5) ꢃ) is sig-
nificantly longer than in other pyridines of the same type, al-
À
though the N(1) C(1) bond is 1.340(5) ꢃ, which is compara-
ble to 2,4,6-triphenyl-substituted pyridines. Also, the
tempted cyclometalation with iridium, no additional NaOAc
is needed, as two acetate groups are already incorporated
À
carbon carbon bond distances alternate much more. Conse-
into the palladium precursor. Moreover, PdACTHGNUTRENNU(G OAc)2 has been
À
quently, the C(2) C(3) bond in rac-12 is shorter than ob-
proven to be a more preferred starting material instead of
Li2PdCl4 since the resulting acetato-bridged dimers, unlike
chloro-bridged dimers, are soluble in common organic
solvents.[19]
served for rac-10 (1.383(6) vs. 1.412(3) ꢃ), whereas the
C(3) C(4) bond is longer (1.435(5) vs. 1.400(3) ꢃ). Also, the
À
C(1)-N(1)-C(13) angle of 117.8(3)8 is considerably smaller
for rac-12 compared to rac-10. Apparently, the unsaturated
bridge in rac-12 has a large effect on the bond lengths of the
nitrogen heterocycle, whereas the distance of the naphthyl
moiety to the heterocyclic moiety appeared to be unaffected
À
In general, C H activation reactions of pyridine-based li-
gands with Pd
A
the presence of an alcohol as solvent, typically MeOH, and
often products are generated in high yields. Unfortunately,
À
with a C(1) C(15) bond length of 1.519(6) ꢃ.
under these conditions, the treatment of PdACTHGNUTRENNU(G OAc)2 with rac-
10 and rac-12 did not give any of the preferred cyclopalla-
dated products. It turned out that a large amount of palladi-
um black is formed within one hour after starting the reac-
tion. Attempts to add the pyridine derivatives dropwise to
the palladium precursor at first appeared promising, but the
reaction mixtures turned black again, overnight, without for-
mation of detectable amounts of products. We, therefore,
carried out the reactions in CH2Cl2, which turned out to be
the solvent of choice in our case.
Reactivity of phosphinines and pyridines towards IrIII and
Pd : Phosphinines 1 and 3 undergo rapid base-assisted C H
activation using [Cp*IrCl2]2 as a metal precursor in the pres-
ence of NaOAc. Both products could recently be character-
ized by single crystal X-ray diffraction and represent the
first examples of IrIII complexes containing cyclometalated
phosphinine ligands (Figure 2, Scheme 5).[5–7] Interestingly,
II
À
A solution of 2 in CH2Cl2 was added dropwise at room
temperature to a solution of PdACHTUNRGTNEUNG(OAc)2 and stirred, over-
night. After 16 h no palladium black was observed and the
CH2Cl2 was exchanged for CD2Cl2 in order to follow the re-
1
action over time by means of H NMR spectroscopy. Analy-
1
À
sis of the H NMR spectrum did not reveal any C H activa-
tion products. The reaction mixture was transferred to
a Young NMR tube and placed in an oil bath at 808C.
Within 6 days all 2,4,6-triphenylpyridine (2) was consumed
and the reaction mixture was filtered over a short pad of
silica to remove any residual palladium black formed during
the course of the reaction. The product was obtained almost
1
quantitatively as a yellow powder. Based on the H NMR
data we anticipated that cyclopalladation occurred according
to Scheme 6.
Scheme 5. Treatment of 1, 3, 2, and 10 with [Cp*IrCl2]2 in the presence of
NaOAc.
Bright yellow crystals of the product, suitable for X-ray
diffraction, were obtained by slow diffusion of n-pentane
into a solution of the crude reaction mixture in THF. Com-
pound 13 crystallizes in the monoclinic space group P21/n
(no.: 14) with one independent molecule in the asymmetric
unit and the molecular structure of 13 in the crystal with se-
neither 2,4,6-triphenylpyridine (2) nor a racemic mixture of
10 can be cyclometalated with [Cp*IrCl2]2 in the presence of
NaOAc, although this reaction is well known for 2-phenyl-
pyridine derivatives.[15,16] The same behavior was observed
for the benzo(h)quinoline rac-12. We assume that the ob-
served lack of reactivity can be attributed purely to steric
factors. As already illustrated in Figure 4 and in contrast to
2,4,6-triarylphosphinines, the more localized nitrogen lone
pair in 2,4,6-triarylpyridines is much better shielded by the
phenyl groups in 2- and 6-position, reducing its accessibility
for s-coordination towards a metal center considerably.
We, therefore, decided to investigate the reactivity of the
pyridine derivatives 2, rac-10 and rac-12 towards PdII com-
plexes as they are sterically less demanding than [Cp*IrCl2]2
Scheme 6. Cyclopalladation of 2 towards Pd-acetate dimer 13.
14462
www.chemeurj.org
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 14458 – 14469

2,4,6-Triarylphosphinines versus 2,4,6-Triarylpyridines
FULL PAPER
Scheme 7. Cyclopalladation of rac-10 and rac-12 towards Pd-acetate
dimers 14 and 15, respectively.
ed yield). The compound crystallizes in the monoclinic
space group C2/c and the molecular structure with selected
bond lengths is shown in Figure 8. The X-ray crystal struc-
ture unambiguously confirms the cyclometalation of rac-10
by PdII with elimination of HOAc and formation of 14. Simi-
lar to coordination compound 13, the molecular structure of
14 in the crystal shows the two acetate moieties bridging
both palladium atoms. Furthermore, both nitrogen heterocy-
Figure 7. Molecular structure of 13 determined by X-ray crystallography.
Displacement ellipsoids are shown at the 50% probability level. Com-
pound 13 co-crystallized with one tetrahydrofuran molecule, which is
À
omitted for clarity. Selected bond lengths [ꢃ]: Pd(1) N(1): 2.0572(17),
À
À
À
Pd(1) C(71):
1.954(2),
À
Pd(1) O(13):
2.1650(16),
Pd(1) O(24):
À
À
2.0400(15), Pd(2) N(2): 2.0459(17), Pd(2) C(72): 1.952(2), Pd(2) O(14):
À
À
2.1497(16), Pd(2) O(23): 2.0299(16). For comparison, the Pd(1) Pd(2)
distance in 13 is 2.9290(2) ꢃ. Lower view: rod representations of the mo-
lecular structure of 13. Left: top view; right: front view.
lected bond lengths (Figure 7). As observed for related com-
pounds, complex 13 exhibits a dimeric structure in which the
two palladium centers are bridged by the two acetates in
such a way that the two halves of the molecule are folded
back on each other, like a closed book.
Similar to the cyclopalladation of 2,4,6-triphenylpyridine,
the atropisomeric pyridine derivative rac-10 was treated
À
with PdACHTUNGTRENNUNG(OAc)2 in CH2Cl2. In contrast to 2, however, the C
H activation of rac-10 proceeds much faster and complete
conversion to coordination compound 14 was already ob-
served after 2.5 days at 808C (Scheme 7, left side). We attri-
bute this enhanced reactivity to the aryl-substituent in 2-po-
sition of the heterocyclic framework, which is already ori-
ented in the plane of the nitrogen heterocycle due to the
C2H4 bridge. This results in an improved accessibility by the
Figure 8. Molecular structure of 14 determined by X-ray crystallography.
Displacement ellipsoids are shown at the 50% probability level. Com-
pound 14 co-crystallized with five tetrahydrofuran molecules, which are
À
omitted for clarity. Selected bond lengths [ꢃ]: Pd(1) N(1): 2.071(4),
À
metal center for the subsequent C H activation reaction.
À
À
À
Pd(1) C(11): 1.948(5), Pd(1) O(1): 2.151(3), Pd(1) O(2): 2.043(4). For
Bright yellow crystals of the product 14, suitable for
single crystal X-ray diffraction, were obtained by slow diffu-
sion of n-pentane into a solution of 14 in THF (70% isolat-
À
comparison, the Pd(1) Pd(1i) distance in 14 is 2.8985(11) ꢃ. Lower view:
rod representations of the molecular structure of 14. Left: top view;
right: front view.
Chem. Eur. J. 2013, 19, 14458 – 14469
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
14463

C. Mꢁller et al.
cles are coordinated to consecutive palladium atoms in op-
posite fashion.
The presence of Pd···Pd interactions can thus not be ex-
AHCTUNGTRENNUNG
cluded.[20,21]
À
In this way both nitrogen atoms are as far apart as possi-
ble, which ultimately results in the fact that the naphthyl
groups in 6-position of the ligand do not interfere with each
other.
Having now explored the C H activation of the pyridine
derivatives 2, rac-10 and rac-12 by PdII, we started to investi-
gate the reactivity of phosphinines 1 and rac-3 towards Pd-
AHCTUNGTRENNUNG
Following the same synthetic procedure described for 13
and 14, the palladium(II) acetate dimer 15 could be isolated
in 40% yield (Scheme 7, right side). Bright yellow crystals
of 15 suitable for single crystal X-ray diffraction were ob-
tained by slow diffusion of Et2O into a solution of 15 in
CH2Cl2. The compound crystallizes in the monoclinic space
group C2/c and the molecular structure with selected bond
lengths is shown in Figure 9. This type of structure had also
been observed in a crystal structure determination[17d] of the
unfunctionalized and much less bulky benzo[h]quinoline.
The Pd···Pd distances in the Pd-acetate dimers 13, 14, and
15 are 2.9290(5), 2.8985(11) and 2.9066(8) ꢃ, respectively.
phosphinine (1) with Pd
(OAc)2 in toluene at À788C.[22] The
authors postulated the oxidation of excess 1 by PdII under
formation of the polymeric Pd0 species [Pd(1)2]n, rather than
À
the formation of a C H activated product (Scheme 8).
Scheme 8. Postulated pathway by Reetz et al.[22] for the formation of
polymeric species [Pd(1)2]n starting from 2,4,6-triphenylphosphinine (1)
and PdACTHNURTGNENG(U OAc)2.
Unsatisfied with these results we repeated the treatment
of 2,4,6-triphenylphosphinine (1) with only 1 equiv of Pd-
(OAc)2 in CH2Cl2 at À788C. After warming the reaction
mixture to room temperature, several resonance between
d=15–50 ppm were detected by 31P{1H} NMR spectroscopy.
However, heating the sample at 808C for 21 h led to the de-
tection of a single resonance in the 31P{1H} NMR at d=
32.6 ppm. This chemical shift at rather high field is neither
typical for l3-phosphinines nor for their corresponding
metal complexes, and is in clear contrast to the observations
made above.[23] Crystals of the reaction product suitable for
X-ray diffraction were obtained by slow diffusion of n-pen-
tane into a solution of crude 16 in THF. The crystallographic
À
characterization unambiguously reveals that no C H activa-
tion of 2,4,6-triphenylphosphinine had occurred. Instead the
cofacial oxidative coupling product 16 had been formed
quantitatively (Scheme 9) and the molecular structure of 16
in the crystal is depicted in Figure 10.
The cofacial oxidative coupling of
1
by [Cu-
AHCTNUGTER(GNNNU CH3CN)4]ClO4 towards 16 and its crystallographic charac-
terization has already been reported by Kojima and Matsu-
da.[24] The molecular structure of 16 in the crystal is essen-
tially identical compared to the one reported in the litera-
À
ture, showing the very long C(12) C(51) bond distance of
À
1.6543(18) ꢃ for a C C single bond. However, the observed
bond lengths and angles differ slightly from the literature
data and selected bond lengths are therefore listed in
Figure 10. The authors claimed that the oxidative coupling
Figure 9. Molecular structure of 15 determined by X-ray crystallography.
Displacement ellipsoids are shown at the 50% probability level. Com-
pound 15 co-crystallized with one dichloromethane molecule, which is
À
required both the presence of ClO4 and CuI. An oxidation
À
of CuI-coordinated 2,4,6-triphenylphosphinine by ClO4
+
À
omitted for clarity. Selected bond lengths [ꢃ]: Pd(2) N(1): 2.076(4),
C
under formation of the radical cation [1] and subsequent
À
À
À
Pd(2) C(11): 1.951(5), Pd(2) O(2): 2.144(4), Pd(2) O(3): 2.046(3),
coupling was proposed.
À
À
À
Pd(1) N(2): 2.078(4), Pd(1) C(46): 1.938(5), Pd(1) O(1): 2.053(4),
We could now demonstrate for the first time that this oxi-
dative coupling reaction is seemingly a more general reac-
tion of 2,4,6-triarylphosphinines than initially believed, as
À
À
Pd(1) O(4): 2.130(3). For comparison, the Pd(1) Pd(2) distance in 15 is
2.9066(8) ꢃ. Lower view: rod representations of the molecular structure
of 15. Left: top view; right: front view.
14464
www.chemeurj.org
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 14458 – 14469

2,4,6-Triarylphosphinines versus 2,4,6-Triarylpyridines
FULL PAPER
Scheme 9. Treatment of 1 with PdACHTUNGTRENUNG(OAc)2 in CH2Cl2 leading to the cofa-
cial oxidative coupling product 16 (newly formed bonds are shown in
gray). The necessary alignment of two phosphinines to form the reaction
product is depicted in the lower right.
the quantitative formation of 16 from 1 can apparently also
be accomplished by Pd
can also be formed quantitatively from 1 by treatment with
Au(OAc)3. Interestingly, treatment of the atropisomeric
phosphinine rac-3 with 1 equiv of Pd(OAc)2 and under the
same reaction conditions as described for phosphinine 1
led again to the formation of main product
ACHTUNGERTN(NUNG OAc)2. Moreover, we found that 16
ACHTUNGTRENNUNG
Figure 10. Molecular structure of 16 in the crystal. Only one independent
molecule is shown. Top view: rod representation resembling the whole
molecule. Bottom view: part of the molecular structure of 16 in the crys-
tal showing only the cage-type structure. Displacement ellipsoids are
shown at the 50% probability level. The phenyl groups have been omit-
ted for clarity and only the ipso-C atom (C61, C62, C181, 182, C121,
AHCTUNGTRENNUNG
a
G
signals at d=24.4 and 24.5 ppm are indeed in the same
region as observed for 16. We, therefore, propose the almost
quantitative formation of the oxidative coupling product 17,
according to Scheme 10.
À
C122) are shown. Selected bond lengths [ꢃ]: P(1) O(1): 1.6257(10);
À
À
À
P(1) O(11): 1.4607(10); P(2) O(1): 1.6284(10); P(2) O(12): 1.4641(10);
À
À
À
P(1) C(11): 1.8248(13); P(1) C(51): 1.8797(13); P(2) C(52): 1.8867(13);
À
À
À
P(2) C(12): 1.8354(14); C(11) C(52): 1.6128(18); C(12) C(51):
The two signals (1:1) observed in the 31P{1H} NMR spec-
trum might originate from the presence of diastereomers.
Unfortunately, no crystallographic characterization of the
product could be performed yet in order to determine also
the absolute configurations of the 2-methylnaphthyl-groups.
In accordance to the observations by Kojima and Matsu-
da, the question of the origin of water and oxygen, which is
apparently necessary for the formation of 16 and 17, howev-
er, is elusive at the moment. Because the results described
here are completely reproducible, traces of water and
1.6543(18); C(41) C(42): 1.5415(18).
À
oxygen in the solvents or the metal precursors might be re-
sponsible for the outcome of the reactions.
Conclusion
Intrigued by investigating the differences between aromatic
phosphorus and nitrogen heterocycles having otherwise
identical substitution patterns we started to synthesize the
novel atropisomeric pyridine derivative rac-10 by treatment
of the corresponding pyrylium salt with aqueous NH3 solu-
tion. By making use of DFT calculations the rotational bar-
rier of the axially chiral pyridine derivative was estimated
and we found that 10 has a considerably lower energy barri-
er for internal rotation compared to its phosphorus analogue
3. However, the presence of the two enantiomers could be
confirmed by means of chiral analytical HPLC analysis and
by protonation experiments with a chiral acid. Compound
Scheme 10. Cofacial oxidative coupling of rac-3 by PdACTHNUTRGNEUNG(OAc)2.
Chem. Eur. J. 2013, 19, 14458 – 14469
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
14465

C. Mꢁller et al.
1.75 (s, 3H, quinolinium-CH3), 1.76 (s, 3H, quinolinium-CH3), 1.76–1.84
(m, 2ꢄ1H, partly overlapping with CH3-groups), 1.86 (t, 2ꢄ1H, J=
4.5 Hz), 1.93 (d, 2ꢄ1H, J=14.5 Hz), 2.14 (dt, 2ꢄ1H, J=18.0, 3.9 Hz),
2.20–2.30 (m, 2ꢄ1H), 2.40 (s, 3H, quinolinium-CH3), 2.42 (s, 3H, quinoli-
nium-CH3), 2.46 (d, 2ꢄ1H, J=14.5 Hz), 2.75–2.85 (m, 2ꢄ2H), 2.92–3.02
(m, 2ꢄ2H), 7.19–7.24 (m, 2ꢄ1H), 7.27–7.39 (m, 2ꢄ3H), 7.43–7.53 (m,
2ꢄ4H), 7.53–7.66 (m, 2ꢄ4H), 7.92–7.96 (m, 2ꢄ1H), 7.98 (d, 2ꢄ1H, J=
8.4 Hz), 8.59–8.72 ppm (m, 2ꢄ1H); 13C NMR (100.63 MHz, CD2Cl2,
258C): d=16.75 (2C), 19.90 (2C, camphor), 20.08 (2C), 20.61 (camphor),
20.66 (camphor), 24.67 (2C, camphor), 26.37 (2C), 27.16 (2C, camphor),
27.77 (2C), 42.96 (2C, camphor), 43.04 (2C, camphor), 46.85 (2C, cam-
phor), 47.72 (2C, camphor), 58.44 (2C, camphor), 124.12, 124.18, 125.81,
127.58, 127.87, 127.90, 128.37, 128.50, 128.73, 128.89, 128.96, 129.19,
129.36, 129.41, 129.43, 129.67, 130.50, 130.54, 132.02, 132.07, 132.21,
132.23, 132.57, 132.62, 134.28, 134.36, 134.55, 134.65, 136.24, 136.29,
136.37, 136.68, 139.58, 147.48, 147.88, 151.07, 151.20, 157.62, 216.47 ppm
(2C, camphor); elemental analysis calcd (%) for C41H41NO4S·1/3CH2Cl2
(672.14): C 73.86, H 6.25, N 2.08; found: C 74.40, H 5.94, N 2.00.
rac-10 could further be dehydrogenated to the benzo(h)qui-
noline derivative rac-12 by treatment with DDQ. This con-
version failed for the phosphorus analogue rac-3. Interest-
À
ingly, although 2,4,6-triarylphosphinines undergo facile C H
activation with [Cp*IrCl2]2 in the presence of NaOAc, no
ortho-metalation was achieved for the pyridine derivatives
2, rac-10 and rac-12. On the other hand, the latter ones can
be selectively ortho-metalated with PdACTHNUTRGENUG(N OAc)2, leading to the
acetate-bridged dimeric species 13, 14 and 15, which could
be unambiguously confirmed by means of X-ray crystal
structure analysis. The treatment of phosphinines with Pd-
ACHTUNGTRENNUNG(OAc)2 led instead to the formation of the unusual cofacial
oxidative coupling products 16 and 17, which consist of
a phosphorus-containing cage structure. This oxidative cou-
pling reaction seems to be a general reaction of 2,4,6-triaryl-
phosphinines as a nearly quantitative formation of the cage
compound cannot only be achieved with different metal
precursors but also with substituted 2,4,6-triarylphosphinine
derivatives.
The results presented here nicely show the difference in
reactivity between aromatic phosphorus and nitrogen het-
erocycles, having otherwise identical structures.
rac-3-Methyl-2-(2-methylnaphthalen-1-yl)-4-phenylbenzoquinoline (12):
Toluene (5.0 mL) was added to a mixture of rac-5 (250.0 mg, 0.61 mmol,
1.0 equiv) and DDQ (275.8 mg, 1.21 mmol, 2.0 equiv) under an argon at-
mosphere. The dark red reaction mixture was refluxed for 19 h. The reac-
tion mixture was allowed to cool to room temperature and the black pre-
cipitate was filtered off. Column chromatography of the filtrate (Si, h=
3.5 cm, w=5.0 cm, n-hex/EtOAc=7:1) gave the product as a pink fluffy
solid (176 mg, 0.43 mmol, 71%). Colorless platelet-like crystals suitable
for X-ray diffraction were obtained from a concentrated EtOH solution
(10 mg, 1.5 mL). M.p.: 95–968C; 1H NMR (400.16 MHz, CD2Cl2, 258C):
d=1.89 (s, 3H, CH3), 2.28 (s, 3H, CH3), 7.25–7.30 (m, 1H), 7.32–7.37 (m,
1H), 7.40–7.49 (m, 4H), 7.51–7.57 (m, 2H), 7.58–7.71 (m, 4H), 7.75 (d,
1H, J=8.8 Hz), 7.87–7.97 (m, 3H), 9.19–9.24 ppm (m, 1H); 13C NMR
(100.63 MHz, CD2Cl2, 258C): d=17.13, 20.30, 124.20, 124.98, 125.20,
125.46, 125.66, 126.76, 127.31, 127,68, 128.00, 128.24, 128.29, 128.31,
128.52, 129.09, 129.11, 129.31, 129.54, 129.93, 129.97, 132.13, 132.59,
132.63, 133.49, 133.88, 137.87, 138.21, 145.16, 148.28, 158.57 ppm; elemen-
tal analysis calcd (%) for C31H23N (409.52): C 90.92, H 5.66, N 3.42;
found: C 90.00, H 5.80, N 3.39.
Experimental Section
General: 1H and 31P{1H} NMR spectra were recorded by using a Varian
Mercury 200 or 400 MHz spectrometer. HPLC analysis was performed
by using HPLC equipment that consisted of a Shimadzu LC-20AD
pump, a Shimadzu SPD-20A prominence UV/Vis detector, a Shimadzu
SIL-20A HT prominence autosampler, and a CTO-20AC prominence
column oven. Column and analysis specifications: Chiralpak IC (250ꢄ
4.6 mm, particle size: 5 mm, purchased from Daicel), eluent=n-hexane/2-
À
General procedure for the attempted C H activation of 2 and 10 with
propanol (99:1), column temperature=258C, flow-rate=1.0 mLminÀ1
,
[Cp*IrCl2]2: A mixture of [Cp*IrCl2]2 (31.9 mg, 0.040 mmol, 1.0 equiv),
pyridine (0.080 mmol, 2.0 equiv), and NaOAc (6.6 mg, 0.080 mmol,
2.0 equiv) was suspended in CD2Cl2 (1.5 mL) and heated to 808C for
2.5 days under argon. The yellow reaction mixture was analyzed by
means of 1H NMR spectroscopy at regular time intervals. No signals of
the cyclometalated product could be detected.
P=32 bar, l=254 nm (UV detector), injection volume=10 mL. HPLC
sample preparation: 5 (0.7 mg) in n-hexane (1.5 mL).
rac-3-Methyl-2-(2-methylnaphthalen-1-yl)-4-phenyl-5,6-dihydrobenzoqui-
noline (10): EtOH (10 mL) was added to rac-4 (100.0 mg, 0.200 mmol,
1.0 equiv) and the yellow suspension turned orange-red upon slow addi-
tion of aqueous ammonia (25%, 10.0 mL). The mixture was refluxed for
3 h resulting in a pale yellow suspension. The off-white solid was filtered
off, redissolved in CH2Cl2, dried over MgSO4 and concentrated to a pale
yellow fluffy solid (81 mg, 0.197 mmol, 81%). Colorless platelet-like crys-
tals suitable for X-ray diffraction were obtained after slow evaporation of
EtOH from a concentrated solution (20 mg, 1.5 mL). M.p.: 170–1718C;
1H NMR (400.16 MHz, CD2Cl2, 258C): d=1.67 (s, 3H, CH3), 2.28 (s, 3H,
CH3), 2.66–2.77 (m, 2H, -CH2-), 2.84–2.95 (m, 2H, -CH2-), 7.19–7.59 (m,
12H), 7.86 (d, 1H, J=8.4 Hz), 7.89 (d, 1H, J=8.2 Hz), 8.18–8.22 ppm
(m, 1H); 13C NMR (100.63 MHz, CD2Cl2, 258C): d=16.75, 20.17, 26.32,
28.48, 125.28, 125.52, 125.72, 126.52, 127.21, 127,91, 127.92 (shoulder),
128.37, 129.03, 129.04 (shoulder) 129.12, 129.20, 129.82, 132.53, 132.61,
133.86, 135.59, 137.63, 138.36, 138.97, 149.68, 150.51, 156.10 ppm; elemen-
tal analysis calcd (%) for C31H25N (411.54): C 90.47, H 6.12, N 3.40;
found: C 89.52, H 6.54, N 3.19.
À
General procedure for the C H activation of 2, 10, and 12 with Pd-
(OAc)2:
A
solution of pyridine (0.049 mmol, 1.0 equiv) in CH2Cl2
(2.0 mL) was added dropwise under an argon atmosphere to a solution of
Pd(OAc)2 (11.0 mg, 0.049 mmol, 1.0 equiv) in CH2Cl2 (2.0 mL) at room
AHCTUNGTRENNUNG
temperature. After being stirred, overnight, all volatiles were removed in
vacuo and the reaction mixture was redissolved in CD2Cl2 (0.5 mL),
transferred to a Young NMR tube and heated to 808C for 2.5 days (com-
pounds 10 and 12) or 6 days (compound 2). During this time the dark re-
1
action mixture was analyzed by means of H NMR spectroscopy at regu-
lar time intervals. When completed, the reaction mixtures were filtered
over a short silica pad, washed, and concentrated to obtain yellow solids,
which were crystallized accordingly. Isolated yields:
2 (4.4 mg, 4.7ꢄ
10À3 mmol, 19%), 10 (19.6 mg, 1.7ꢄ10À2 mmol, 70%), 12 (11.2 mg, 9.7ꢄ
10À3 mmol, 40%).
2,4,6-Triphenylpyridine palladium(II) acetate dimer (13): 1H NMR
(400.16 MHz, CD2Cl2, 258C): d=1.30 (s, 2ꢄ3H), 6.81 (dt, 2ꢄ1H, J=7.4,
1.4 Hz), 6.86 (d, 2ꢄ1H, J=2.0 Hz), 6.92 (dd, 2ꢄ1H, J=7.8, 1.2 Hz),
6.96–7.02 (m, 2ꢄ2H), 3.37–7.48 (m, 2ꢄ4H), 7.49 (d, 2ꢄ1H, J=1.8 Hz),
7.51–7.64 (m, 2ꢄ5H), 7.68–7.72 ppm (m, 2ꢄ2H); 13C NMR (100.63 MHz,
CD2Cl2, 258C): d=22.96 (2C), 114.06 (2C), 121.30 (2C), 123.02 (2C),
123.53 (2C), 127.14 (2ꢄ2C), 127.45 (2C), 127.89 (2ꢄ2C), 182.82 (2ꢄ2C),
129.20 (2ꢄ2C), 129.75 (2ꢄ2C), 131.95 (2C), 137.04 (2C), 139.37 (2C),
145.17 (2C), 149.51 (2C), 150.32 (2C), 161.65 (2C), 165.07 (2C),
178.34 ppm (2C).
3-Methyl-2-(2-methylnaphthalen-1-yl)-4-phenyl-5,6-dihydrobenzoquinoli-
nium-(1S)-camphor-10-sulfonate (diastereomeric mixture) (11): CH2Cl2
(2.0 mL) was added to a mixture of 5 (32.9 mg, 0.080 mmol, 1.0 equiv)
and (1S)-(+)-camphor-10-sulfonic acid (18.6 mg, 0.080 mmol, 1.0 equiv).
The clear yellow solution was stirred for 30 min at room temperature. Af-
terwards all volatiles were removed in vacuo and the product was ob-
tained as a pale yellow solid (51.2 mg, 0.080 mmol, 99%). 1H NMR
(400.16 MHz, CD2Cl2, 258C): d=0.62 (s, 2ꢄ3H, camphor-CH3), 0.86 (s,
2ꢄ3H, camphor-CH3), 1.03–1.19 (m, 4H), 1.63 (d, 2ꢄ1H, J=18.0 Hz),
14466
www.chemeurj.org
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 14458 – 14469

2,4,6-Triarylphosphinines versus 2,4,6-Triarylpyridines
FULL PAPER
rac-3-Methyl-2-(2-methylnaphthalen-1-yl)-4-phenyl-5,6-dihydrobenzoqui-
noline palladium(II) acetate dimer (14): 1H NMR (400.16 MHz, CD2Cl2,
258C): d=0.36 (s, 2ꢄ3H), 1.34 (s, 2ꢄ3H), 2.47–2.55 (m, 2ꢄ2H), 2.58 (s,
2ꢄ3H), 2.67–2.88 (m, 2ꢄ2H), 6.72 (t, 2ꢄ1H, J=6.6 Hz), 6.85 (t, 2ꢄ1H,
J=7.4 Hz), 7.18 (dt, 2ꢄ1H, J=7.0, 1.4 Hz), 7.23 (dt, 2ꢄ1H, J=7.6,
1.4 Hz), 7.41–7.62 (m, 2ꢄ8H), 7.73 (d, 2ꢄ1H, J=8.2 Hz), 7.82 (d, 2ꢄ1H,
J=8.4 Hz), 7.85 ppm (d, 2ꢄ1H, J=7.8 Hz); 13C NMR (100.63 MHz,
CD2Cl2, 258C): d=17.90 (2C), 21.32 (2C), 22.62 (2C), 25.65 (2C), 27.90
(2C), 123.19 (2C), 125.20 (2C), 126.41 (2C), 126.59 (2C), 126.60 (2C),
127.93 (2C), 128.16 (2C), 128.35 (2C), 128.37 (2C), 128.67 (2C), 128.87
(2C), 129.17 (2C), 129.31 (2C), 129.84 (2C), 130.46 (2C), 131.61 (2C),
132.39 (2C), 132.50 (2C), 134.65 (2C), 135.53 (2C), 135.75 (2C), 138.12
(2C), 143.50 (2C), 149.79 (2C), 150.91 (2C), 156.01 (2C), 160.00 (2C),
with one restraint. R1 =0.063 for 10003 reflections with I>2s(I) and
wR2 =0.174 for 13918 reflections, S=1.080, residual electron density was
between À0.36 and 0.56 eꢃÀ3. Geometry calculations and checks for
higher symmetry were performed with the PLATON program.[28] Friedel
opposites were not merged. The absolute structure could not be deter-
mined reliably due to a high error value (Flack x parameter=À1.2298
with su 3.9249).
X-ray crystal structure determination of 13: Crystals suitable for X-ray
diffraction were obtained by slow diffusion of n-pentane into a solution
of 13 in THF. Crystallographic data: C50H38N2O4Pd2·C4H8O; Fw=
1015.73; 0.47ꢄ0.29ꢄ0.06 mm3; yellow platelet, monoclinic; P21/n (no.:
14); a=15.3562(6), b=15.1161(4), c=19.7279(6) ꢃ; a=90, b=
104.628(6), g=908; V=4430.9(3) ꢃ3; Z=4; 1x =1.523 gcmÀ3
;
m=
0.864 mmÀ1; 79341 reflections were measured by using a Bruker Kappa
ApexII diffractometer with sealed tube and Triumph monochromator
(l=0.71073 ꢃ) at a temperature of 150(2) K up to a resolution of (sinq/
l)max =0.65 ꢃÀ1. Intensity data were integrated with the Eval15 soft-
ware.[29] Absorption correction and scaling was performed with
SADABS[26] (correction range 0.63–0.75); 10163 reflections were unique
(Rint =0.027), of which 8605 were observed [I>2s(I)]. The structure was
solved with the program SHELXS-97.[27] Least-squares refinement was
performed with SHELXL-2013[27] against F2 of all reflections. Non-hydro-
gen atoms were refined freely with anisotropic displacement parameters.
Hydrogen atoms of the metal complex were located in difference Fourier
maps and hydrogen atoms of the solvent molecule were introduced in
calculated positions. All hydrogen atoms were refined with a riding
model. The THF solvent molecule was refined with a disorder model;
580 parameters were refined with 20 restraints (distances, angles, and dis-
placement parameters in the disordered THF). R1/wR2 [I>2s(I)]: 0.0255/
0.0641. R1/wR2 [all refl.]: 0.0341/0.0679; S=1.033. Residual electron den-
sity between À0.58 and 0.55 eꢃÀ3. Geometry calculations and checking
for higher symmetry were performed with the PLATON program.[28]
179.79 ppm
(2C);
elemental
analysis
calcd
(%)
for
C66H56N2O4Pd2·1.5CH2Cl2 (1281.16): C 63.27, H 4.64, N 2.19; found: C
63.80, H 5.14, N 2.32.
rac-3-Methyl-2-(2-methylnaphthalen-1-yl)-4-phenylbenzoquinoline palla-
dium(II) acetate dimer (15): 1H NMR (400.16 MHz, CD2Cl2, 258C): d=
0.46 (s, 2ꢄ3H), 1.58 (s, 2ꢄ3H), 2.43 (s, 2ꢄ3H), 6.85 (d, 2ꢄ1H, J=
9.0 Hz), 6.94 (dt, 2ꢄ1H, J=6.6, 1.0 Hz), 7.00 (d, 2ꢄ1H, J=7.4 Hz), 7.04
(d, 2ꢄ1H, J=9.0 Hz), 7.16 (dt, 2ꢄ1H, J=6.8, 1.6 Hz), 7.51–7.63 (m, 2ꢄ
6H), 7.66–7.75 (m, 2ꢄ2H), 7.88 (d, 2ꢄ2H, J=8.0 Hz), 7.92 ppm (d, 2ꢄ
2H, J=8.2 Hz); 13C NMR (100.63 MHz, CD2Cl2, 258C): d=18.42 (2C),
21.19 (2C), 22.69 (2C), 121.28 (2C), 122.10 (2C), 124.75 (2C), 125.00
(2C), 125.43 (2C), 126.63 (2C), 126.67 (2C), 128.56 (2C), 128.75 (2C),
129.02 (shoulder, 2ꢄ2C), 129.04 (2ꢄ2C), 129.30 (2C), 129.53 (2C), 129.92
(2C), 130.04 (2C), 130.69 (2C), 131.64 (2C), 132.39 (2C), 132.47 (2C),
134.77 (2C), 135.61 (2C), 137.42 (2C), 140.90 (2C), 147.67 (2C), 149.21
(2C), 152.81 (2C), 159.05 (2C), 180.36 ppm (2C); elemental analysis calcd
(%) for C66H52N2O4Pd2·1.5CH2Cl2 (1276.63): C 63.47, H 4.34, N 2.19;
found: C 64.11, H 4.65, N 2.13.
X-ray crystal structure determination of 10: Crystals suitable for X-ray
diffraction were obtained by slow evaporation of EtOH from a concen-
trated solution of 10. Crystallographic data: C31H25N; Fw=411.52; 0.16ꢄ
0.08ꢄ0.02 mm3; colorless platelet, monoclinic; P21/n (no.: 14); a=
11.280(3), b=11.164(3), c=17.476(4) ꢃ; a=90, b=90.913(6), g=908;
V=2200.5(10) ꢃ3; Z=4; 1x =1.242 gcmÀ3; m=0.541 mmÀ1; 13862 reflec-
tions were measured on a Bruker Proteum diffractometer with a rotating
anode and Helios optics (l=1.54184 ꢃ) at a temperature of 100(2) K up
to a resolution of (sinq/l)max =0.59 ꢃÀ1. Intensity data were integrated
with the Saint software.[25] Absorption correction and scaling was per-
formed with SADABS[26] (correction range 0.59–0.75). 3687 reflections
were unique (Rint =0.032), of which 3599 were observed [I>2s(I)]. The
structure was solved with the program SHELXS-97.[27] Least-squares re-
finement was performed SHELXL-97[27] against F2 of all reflections.
Non-hydrogen atoms were refined freely with anisotropic displacement
parameters. Hydrogen atoms were located in difference Fourier maps
and refined with a riding model; 291 parameters were refined with no re-
straints. R1/wR2 [I>2s(I)]: 0.053/0.1419. R1/wR2 [all refl.]: 0.0561/0.1424;
S=1.126. Residual electron density between À1.25 and 0.55 eꢃÀ3. Geom-
etry calculations and checking for higher symmetry were performed with
the PLATON program.[28]
X-ray crystal structure determination of 14: Crystals suitable for X-ray
diffraction were obtained by slow diffusion of n-pentane into a solution
of 14 in THF. Crystallographic data: C33H27NO2Pd·2.5(C4H8O); Fw=
756.22; 0.12ꢄ0.26ꢄ0.34 mm3; yellow block, monoclinic; C2/c; a=
18.192(4), b=21.406(4), c=19.668(4) ꢃ; a=90, b=106.11(3), g=908;
V=7359(3) ꢃ3; Z=8; 1x =1.365 gcmÀ3; m=0.549 mmÀ1; 18880 reflections
were measured by using a Stoe IPDS 2T diffractometer with a rotating
anode (MoKa radiation; l=0.71073 ꢃ) up to a resolution of (sinq/l)max
=
0.64 ꢃÀ1 at a temperature of 220 K. The reflections were corrected for
absorption and scaled on the basis of multiple measured reflections by
using the X-Red program (0.80–0.93 correction range);[30] 6448 reflections
were unique (Rint =0.067). The structures were solved with SHELXS-
97[27] by using direct methods and refined with SHELXL-97[27] on F2 for
all reflections. Non-hydrogen atoms were refined by using anisotropic
displacement parameters. The positions of the hydrogen atoms were cal-
culated for idealized positions; 385 parameters were refined with ten re-
straints. R1 =0.049 for 3903 reflections with I>2s(I) and wR2 =0.124 for
6448 reflections, S=0.868. Residual electron density was between À0.68
and 0.67 eꢃÀ3. Geometry calculations and checks for higher symmetry
were performed with the PLATON program.[28]
X-ray crystal structure determination of 12: Crystals suitable for X-ray
diffraction were obtained by slow evaporation of EtOH from a concen-
trated solution of 12. Crystallographic data: C31H23N; Fw=409.50; 0.12ꢄ
0.22ꢄ0.54 mm3; colorless platelet, monoclinic; P21; a=9.682(3), b=
25.195(8), c=18.494(6) ꢃ; a=90, b=97.726(7), g=908; V=4470 (2) ꢃ3;
Z=8; 1x =1.217 gcmÀ3; m=0.541 mmÀ1; 13918 reflections were measured
by using a Bruker-AXS smart CCD area detector diffractometer (graph-
X-ray crystal structure determination of 15: Crystals suitable for X-ray
diffraction were obtained by slow diffusion of Et2O into a solution of 15
in CH2Cl2. Crystallographic data: C66H50N2O4Pd2·CH2Cl2; Fw=1232.81;
0.09ꢄ0.37ꢄ0.86 mm3; yellow needle, monoclinic; C2/c; a=31.049(6), b=
14.721(3), c=26.355(5) ꢃ; a=90, b=105.88(3), g=908; V=11587(4) ꢃ3;
Z=8; 1x =1.413 gcmÀ3; m=0.763 mmÀ1; 27134 reflections were measured
by using a Stoe IPDS 2T diffractometer with a rotating anode (MoKa ra-
diation; l=0.71073 ꢃ) up to a resolution of (sinq/l)max =0.64 ꢃÀ1 at
a temperature of 220 K. The reflections were corrected for absorption
and scaled on the basis of multiple measured reflections by using the X-
Red program (0.65–0.88 correction range);[30] 10108 reflections were
unique (Rint =0.069). The structures were solved with SHELXS-97[27] by
using direct methods and refined with SHELXL-97[27] on F2 for all reflec-
tions. Non-hydrogen atoms were refined by using anisotropic displace-
ment parameters. The positions of the hydrogen atoms were calculated
for idealized positions; 694 parameters were refined without restraints.
ite monochromator, l=0.71073 ꢃ) up to a resolution of (sinq/l)max
=
0.61 ꢃÀ1 at a temperature of 100 K. The reflections were corrected for
absorption and scaled on the basis of multiple measured reflections by
using the SADABS program[24] (0.60–0.86 correction range); 13918 re-
flections were unique (Rint =0.050). The structures were solved with
SHELXS-97[27] by using direct methods and refined with SHELXL-97[27]
on F2 for all reflections. Non-hydrogen atoms were refined by using ani-
sotropic displacement parameters. The positions of the hydrogen atoms
were calculated for idealized positions; 1161 parameters were refined
Chem. Eur. J. 2013, 19, 14458 – 14469
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
14467

C. Mꢁller et al.
R1 =0.044 for 5581 reflections with I>2s(I) and wR2 =0.101 for 10108
reflections, S=0.794: Residual electron density was between À0.58 and
0.68 eꢃÀ3. Geometry calculations and checks for higher symmetry were
performed with the PLATON program.[28] One disordered unidentified
solvent molecule were not modeled and the disordered density was taken
into account using the SQUEEZE/PLATON procedure.[31]
Vogt, Dalton Trans. 2007, 5372–5375; e) C. Mꢁller, E. A. Pidko,
A. J. P. M. Staring, M. Lutz, A. L. Spek, R. A. van Santen, D. Vogt,
Chem. Eur. J. 2008, 14, 4899–4905; f) C. Mꢁller, Z. Freixa, M. Lutz,
A. L. Spek, D. Vogt, P. W. N. M. van Leeuwen, Organometallics
2008, 27, 834–838; g) C. Mꢁller, E. A. Pidko, M. Lutz, A. L. Spek,
D. Vogt, Chem. Eur. J. 2008, 14, 8803–88078; h) C. Mꢁller, D. Vogt,
C. R. Chim. 2010, 13, 1127–1143.
General procedure for the cofacial oxidative coupling reaction of 1 and
[4] a) A. Campos Carrasco, E. A. Pidko, A. M. Masdeu-Bultꢆ, M. Lutz,
A. L. Spek, D. Vogt, C. Mꢁller, New J. Chem. 2010, 34, 1547–1550;
b) I. de Krom, L. E. E. Broeckx, M. Lutz, C. Mꢁller, Chem. Eur. J.
2013, 19, 3676–3684.
[5] J. J. M. Weemers, W. N. P. van der Graaff, E. A. Pidko, M. Lutz, C.
Mꢁller, Chem. Eur. J. 2013, 19, 8991–9004.
[6] L. E. E. Broeckx, M. Lutz, D. Vogt, C. Mꢁller, Chem. Commun.
2011, 47, 2003–2005.
[7] C. Mꢁller, L. E. E. Broeckx, I. de Krom, J. J. M. Weemers, Eur. J.
Inorg. Chem. 2013, 187–202.
rac-3 with Pd
in CH2Cl2 (2.0 mL) was added dropwise under an argon atmosphere to
solution of Pd(OAc)2 (11.0 mg, 0.049 mmol, 1.0 equiv) in CH2Cl2
ACHTUNGTRENNUNG(OAc)2: A solution of phosphinine (0.049 mmol, 1.0 equiv)
a
ACHTUNGTRENNUNG
(2.0 mL) at À788C. After being stirred, overnight, all volatiles were re-
moved in vacuo and the reaction mixture was redissolved in CD2Cl2
(0.5 mL), transferred to a Young NMR tube and analyzed with 31P NMR
spectroscopy. The reaction mixture was subsequently heated to 808C for
21 h (compound 16) or 3 h (compound 17). During this time the dark re-
1
action mixture was analyzed by means of H NMR spectroscopy at regu-
lar time intervals. When completed, the reaction mixtures were filtered
over a short silica pad, washed, and concentrated to obtain yellow solids.
Compound 16 was crystallized accordingly. No crystals suitable for X-ray
crystal structure determination could be obtained for compound 17.
[8] a) A. R. Katritzky, S. S. Thind, J. Chem. Soc. Perkin 1 1980, 1895–
1900; b) A. R. Katritzky, A. M. El-Mowafy, G. Musumarra, K. Saki-
zadeh, C. Sana-Ullah, S. M. M. El-Shafie, S. S. Thind, J. Org. Chem.
1981, 46, 3823–3830.
X-ray crystal structure determination of 16: Crystals suitable for X-ray
diffraction were obtained by slow diffusion of n-pentane into a solution
of crude 16 in THF. Crystallographic data: C46H34N3O4P2; Fw=696.67;
0.39ꢄ0.23ꢄ0.09 mm3; colorless platelet, monoclinic; P21/n (no.: 14); a=
23.4819(5), b=13.4807(3), c=24.5608(5) ꢃ; a=90, b=116.124(1), g=
908; V=6980.5(3) ꢃ3; Z=8; 1x =1.326 gcmÀ3; m=0.17 mmÀ1; 118692 re-
flections were measured on a Bruker Kappa ApexII diffractometer with
sealed tube and Triumph monochromator (l=0.71073 ꢃ) at a tempera-
ture of 150(2) K up to a resolution of (sinq/l)max =0.65 ꢃÀ1. Intensity data
were integrated with the Eval15 software.[29] Absorption correction and
scaling was performed with SADABS[26] (correction range 0.70–0.75);
16055 reflections were unique (Rint =0.037), of which 13100 were ob-
served [I>2s(I)]. The structure was solved with the program
SHELXT.[32] Least-squares refinement was performed with SHELXL-
97[27] against F2 of all reflections. Non-hydrogen atoms were refined
freely with anisotropic displacement parameters. All hydrogen atoms
were located in difference Fourier maps. Phenyl hydrogen atoms were re-
fined with a riding model. All other hydrogen atoms were refined freely
with anisotropic displacement parameters; 951 parameters were refined
with no restraints. R1/wR2 [I>2s(I)]: 0.0351/0.0846. R1/wR2 [all refl.]:
0.0476/0.0902; S=1.037. Residual electron density between À0.42 and
0.34 eꢃÀ3. Geometry calculations and checking for higher symmetry
were performed with the PLATON program.[28]
[9] DFT calculations were performed at the B3LYP/6–31G level of
theory using the Gaussian09 program (see ref. [10]). The barrier of
rotation was estimated by scanning the potential energy surface
along the coordinate corresponding to the internal rotation about
À
the Caryl Carylꢇ bond with a step size of 208.
[10] Gaussian09, RevisionA.01, M. J. Frisch, G. W. Trucks, H. B. Schle-
gel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, G. Scalmani, V.
Barone, B. Mennucci, G. A. Petersson, H. Nakatsuji, M. Caricato, X.
Li, H. P. Hratchian, A. F. Izmaylov, J. Bloino, G. Zheng, J. L. Son-
nenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa,
M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, T. Vreven,
J. A. Montgomery, Jr., J. E. Peralta, F. Ogliaro, M. Bearpark, J. J.
Heyd, E. Brothers, K. N. Kudin, V. N. Staroverov, R. Kobayashi, J.
Normand, K. Raghavachari, A. Rendell, J. C. Burant, S. S. Iyengar,
J. Tomasi, M. Cossi, N. Rega, J. M. Millam, M. Klene, J. E. Knox,
J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E.
Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W.
Ochterski, R. L. Martin, K. Morokuma, V. G. Zakrzewski, G. A.
Voth, P. Salvador, J. J. Dannenberg, S. Dapprich, A. D. Daniels, O.
Farkas, J. B. Foresman, J. V. Ortiz, J. Cioslowski, and D. J. Fox,
Gaussian, Inc., Wallingford, CT, 2009.
[11] a) E. F. DiMauro, M. C. Kozlowski, J. Chem. Soc. Perkin Trans.
1 2002, 439–444; b) P. Le Floch, Coord. Chem. Rev. 2006, 250, 627–
681; c) C. Mꢁller, D. Vogt, Dalton Trans. 2007, 5505–5523; d) C.
Mꢁller, D. Vogt, in Phosphorus Chemistry: Catalysis and Material
Science Applications, Vol. 36 (Ed.: M. Peruzzini, L. Gonsalvi), Chap-
ter 6, Springer, Heidelberg, 2011.
CCDC-943657 (10), -945744 (12), -943658 (13), -945745 (14), -945746 (15)
and -943659 (16) contain the supplementary crystallographic data for this
paper. These data can be obtained free of charge from The Cambridge
Crystallographic Data Centre via www.ccdc.cam.ac.uk/data_request/cif.
´
[12] a) J. Ondrꢈcek, J. Novotny, M. Petru, P. Lhotꢈk, J. Kuthan, Acta
Crystallogr. C 1994, 50, 1809–1811; b) L. L. Lv, X.-Q. Huang, Acta
Crystallogr. E 2008, 64, o186.
[13] M. R. Bryce, J. Chem. Soc. Perkin Trans. 1 1985, 1675–1679.
[14] The four independent molecules differ in the orientation of the
phenyl group in 4-position of the heterocycle as well as in the orien-
tation of the methyl-napthyl-group with respect to the 6-membered
nitrogen ring.
Acknowledgements
This research has been funded by The Netherlands Organization for Sci-
entific Research (NWO-Vidi, C.M.). The COST-Action PhoSciNet
(CM0802) is gratefully acknowledged.
[15] D. L. Davies, O. Al-Duaij, J. Fawcett, M. Giardiello, S. T. Hilton,
D. R. Russel, Dalton Trans. 2003, 4132–4138.
[16] a) Y. Boutadla, O. Al-Duaij, D. L. Davies, G. A. Griffith, K. Singh,
Organometallics 2009, 28, 433–440; b) C. Scheeren, F. Maasarani,
A. Hijazi, J.-P. Djukic, M. Pfeffer, Organometallics 2007, 26, 3336–
3345; c) D. L. Davies, S. M. A. Donald, O. Al-Duaij, S. A. Macgre-
gor, M. Pçlleth, J. Am. Chem. Soc. 2006, 128, 4210–4211; d) D. L.
Davies, S. M. A. Donald, O. Al-Duaij, J. Fawcett, G. Little, S. A.
Macgregor, Organometallics 2006, 25, 5976–5978.
[17] a) A. J. Canty, N. J. Minchin, B. W. Skelton, A. H. White, J. Chem.
Soc. Dalton Trans. 1986, 2205–2210; b) A. R. Dick, K. L. Hull, M. S.
Sanford, J. Am. Chem. Soc. 2004, 126, 2300–2301; c) A. R. Dick,
J. W. Kampf, M. S. Sanford, J. Am. Chem. Soc. 2005, 127, 12790–
[1] G. Mꢀrkl, Angew. Chem. 1966, 78, 907–908; Angew. Chem. Int. Ed.
Engl. 1966, 5, 846–847.
[2] A. J. Ashe III, J. Am. Chem. Soc. 1971, 93, 3293–3295.
[3] a) C. Mꢁller, L. Guarrotxena-Lopꢅz, H. Kooijman, A. L. Spek, D.
Vogt, Tetrahedron Lett. 2006, 47, 2017–2020; b) C. Mꢁller, M. Lutz,
A. L. Spek, D. Vogt, J. Chem. Crystallogr. 2006, 36, 869–874; c) C.
Mꢁller, D. Wasserberg, J. J. M. Weemers, E. A. Pidko, S. Hoffmann,
M. Lutz, A. L. Spek, S. C. J. Meskers, R. A. J. Janssen, R. A. van
Santen, D. Vogt, Chem. Eur. J. 2007, 13, 4548–4559; d) C. Mꢁller,
E. A. Pidko, D. Totev, M. Lutz, A. L. Spek, R. A. van Santen, D.
14468
www.chemeurj.org
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 14458 – 14469

2,4,6-Triarylphosphinines versus 2,4,6-Triarylpyridines
FULL PAPER
12791; d) D. C. Powers, D. Benitez, E. Tkatchouk, W. A. Goddar-
d III, T. Ritter, J. Am. Chem. Soc. 2010, 132, 14092–14103; e) Y. Ye,
N. D. Ball, J. W. Kampf, M. S. Sanford, J. Am. Chem. Soc. 2010, 132,
14682–14687.
[24] T. Kojima, Y. Ishioka, Y. Matsuda, Chem. Commun. 2004, 366–367.
[25] Bruker. SAINT-plus, Bruker AXS, Inc., Madison, Wisconsin, 2001.
[26] G. M. Sheldrick, SADABS: Area-Detector Absorption Correction,
University of Gçttingen, Gçttingen, 1999.
[18] P. G. Evans, N. A. Brown, G. J. Clarkson, C. P. Newman, J. P.
Rourke, J. Organomet. Chem. 2006, 691, 1251–1256.
[27] G. M. Sheldrick, Acta Crystallogr. A 2008, 64, 112–122.
[28] A. L. Spek, Acta Crystallogr. D 2009, 65, 148–155.
[29] A. M. M. Schreurs, X. Xian, L. M. J. Kroon-Batenburg, J. Appl.
Crystallogr. 2010, 43, 70–82.
[30] P. Coppens, in Crystallographic Computing (Eds.: F. R. Ahmed,
S. R. Hall, C. P. Huber), Munksgaard, Copenhagen, 1979, pp. 255–
270.
[31] P. v. d. Sluis, A. L. Spek, Acta Crystallogr. A 1990, 46, 194–201.
[32] G. M. Sheldrick, SHELXTL, University of Gçttingen, Gçttingen,
2012.
[19] a) G. E. Hartwell, R. V. Lawrence, M. J. Smas, Chem. Commun.
1970, 912; b) M. A. Gutierrez, G. R. Newkome, J. Selbin, J. Organo-
met. Chem. 1980, 202, 341–350; c) J. Selbin, M. A. Gutierrez, J. Or-
ganomet. Chem. 1983, 246, 95–104; d) J. Selbin, K. Abboud, S. F.
Watkins, M. A. Gutierrez, F. R. Fronczek, J. Organomet. Chem.
1983, 241, 259–268.
[20] M. R. Churchill, Perspect. Struct. Chem. 1970, 3, 91–164.
[21] G. Aullꢆn, S. Alvarez, Chem. Eur. J. 1997, 3, 655–664.
[22] M. T. Reetz, E. Bohres, R. Goddard, M. C. Holthausen, W. Thiel,
Chem. Eur. J. 1999, 5, 2101–2108.
[23] C. Mꢁller, in Phosphorus Ligand Effects in Homogeneous Catalysis:
Design and Synthesis (Eds.: P. C. J. Kamer, P. W. N. M van Leeu-
wen), Wiley-VCH, Weinheim, 2012.
Received: June 25, 2013
Published online: October 2, 2013
Chem. Eur. J. 2013, 19, 14458 – 14469
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
14469